Announcement

Collapse
No announcement yet.

MS2 Idle tuning help

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    MS2 Idle tuning help

    I'm having a bitch of a time getting my fucking car to idle smoothly. I've tried open-loop, closed-loop, and even cracking the throttle plate. But no matter what I do, it surges. Sometimes. It was idling perfect when I parked it last summer, now it's surging ~75% of the time I drive it. The other 25%, it's fine. I'm totally out of ideas. Anyone want to look over my logs/tunes and give me some pointers? Or if you have a good m20 or m30 idle tune, car to share the numbers/settings?

    Access Google Drive with a Google account (for personal use) or Google Workspace account (for business use).


    Access Google Drive with a Google account (for personal use) or Google Workspace account (for business use).

    #2
    Weather has a pretty adverse effect on mine. How are your AFR's? I'm at work so I can't look at it right away.
    1989 325i - 2.7i, Holset H1C, 60lb injectors, whodwho MS-PNP.
    2012 Passat TDI - DD Duty
    2008 GMC Yukon XL Denali - Kiddie hauler/grocery getter

    Comment


      #3
      Originally posted by CorvallisBMW View Post
      I'm having a bitch of a time getting my fucking car to idle smoothly. I've tried open-loop, closed-loop, and even cracking the throttle plate. But no matter what I do, it surges. Sometimes. It was idling perfect when I parked it last summer, now it's surging ~75% of the time I drive it. The other 25%, it's fine. I'm totally out of ideas. Anyone want to look over my logs/tunes and give me some pointers? Or if you have a good m20 or m30 idle tune, car to share the numbers/settings?

      Access Google Drive with a Google account (for personal use) or Google Workspace account (for business use).


      https://drive.google.com/open?id=0B4...nNQbE1OWHRxdGs
      I have a few suggestions from looking at the log, I didn't look at the msq itself yet. First, smooth the idle VE cells so that there aren't large differences between adjacent cells, this worsens hunting. I see some oscillation in the spark advance too and the advance numbers are high in my opinion. I would enable idle advance, set a curve that smoothly drops advance above 900rpm or so, and adds a few degrees below 700. Mine idles at 900rpm and about 9 degrees of advance. I also saw some erroneous accel enrich spikes, you should increase smoothing or decrease sensitivity of the TPS or MAP accel enrich.

      Originally posted by redsubdivisions View Post
      Weather has a pretty adverse effect on mine. How are your AFR's? I'm at work so I can't look at it right away.
      It shouldn't. My idle is steady and solid once it settles down no matter the weather.

      IG @turbovarg
      '91 318is, M20 turbo
      [CoTM: 4-18]
      '94 525iT slicktop, M50B30 + S362SX-E, 600WHP DD or bust
      - updated 1-26

      Comment


        #4
        Originally posted by varg View Post
        I have a few suggestions from looking at the log, I didn't look at the msq itself yet. First, smooth the idle VE cells so that there aren't large differences between adjacent cells, this worsens hunting. I see some oscillation in the spark advance too and the advance numbers are high in my opinion. I would enable idle advance, set a curve that smoothly drops advance above 900rpm or so, and adds a few degrees below 700. Mine idles at 900rpm and about 9 degrees of advance. I also saw some erroneous accel enrich spikes, you should increase smoothing or decrease sensitivity of the TPS or MAP accel enrich.
        OK, thanks for the tips!

        Comment


          #5
          Check your limits for ICV frequency for full open and full closed. Also setup closed loop with Gain of ~250, P=100 and I+D=0

          Check if the car is going into closed loop (CL) when it shouldn't, then adjust the CL parameters as required to ensure it only goes into it when needed. Also make sure it's not jumping out of CL when it makes an adjustment, sometimes the RPMdot (rate of RPM change) can be high enough to break it out of CL (surge).

          Another one you should try is setting up the initial value table and just set it all about 10Hz higher than your typical idle value, then you can fine tune it later (just remember to always have it a few hz higher than the actual idle, because it has to respond to a negative RPMdot)

          Hope that helps some.

          Comment


            #6
            Originally posted by MercyFlush View Post
            Check your limits for ICV frequency for full open and full closed. Also setup closed loop with Gain of ~250, P=100 and I+D=0

            Check if the car is going into closed loop (CL) when it shouldn't, then adjust the CL parameters as required to ensure it only goes into it when needed. Also make sure it's not jumping out of CL when it makes an adjustment, sometimes the RPMdot (rate of RPM change) can be high enough to break it out of CL (surge).

            Another one you should try is setting up the initial value table and just set it all about 10Hz higher than your typical idle value, then you can fine tune it later (just remember to always have it a few hz higher than the actual idle, because it has to respond to a negative RPMdot)

            Hope that helps some.
            When I was tuning my M20, I got the car to idle at operating temp about 100rpm lower than desired, with the IAC closed. I locked timing at 17 degrees and set the engine at about 650rpm with the idle screw.

            Then I did my IACV adjustments. It helps the IACV not have to work as hard and makes's it easier to tune the curve. (I was using VEMS).
            1985 325e 2.8 Turbo VEMS

            Comment


              #7
              Originally posted by MercyFlush View Post
              Check your limits for ICV frequency for full open and full closed. Also setup closed loop with Gain of ~250, P=100 and I+D=0

              Check if the car is going into closed loop (CL) when it shouldn't, then adjust the CL parameters as required to ensure it only goes into it when needed. Also make sure it's not jumping out of CL when it makes an adjustment, sometimes the RPMdot (rate of RPM change) can be high enough to break it out of CL (surge).

              Another one you should try is setting up the initial value table and just set it all about 10Hz higher than your typical idle value, then you can fine tune it later (just remember to always have it a few hz higher than the actual idle, because it has to respond to a negative RPMdot)

              Hope that helps some.
              I messed around with the Active and Inactive duty cycles, and IIRC set them to about 45.0 and 20.0 respectively. THis seemed to help out the idle a bunch. I also pulled timing from the RPM area of the spark table and got it down to a reasonable 10 degrees (it was like 22 degrees before!). At this point it actually seems to be idling very well, with good warmup stability, AFRs, etc. I remember upping my RPMdot a few points, and that seemed to prevent the occasional Accel Enrich spike I was seeing.

              I don't quite understand the "initial value table" things, I don't remember seeing an idle valve freq table anywhere?

              Originally posted by SamE30e View Post
              When I was tuning my M20, I got the car to idle at operating temp about 100rpm lower than desired, with the IAC closed. I locked timing at 17 degrees and set the engine at about 650rpm with the idle screw.

              Then I did my IACV adjustments. It helps the IACV not have to work as hard and makes's it easier to tune the curve. (I was using VEMS).
              Hmm interesting idea, I hadn't thought of splitting the idle duty between the idle screw and ICV. I can see how this would allow the ICV to operate in a smaller range, and possible reduce the chances of hunting/surging since it's only making small adjustments. If the idle suddenly goes to shit one day (as it seems to do occasionally, for no apparent reason) I might try this.

              On additional weird fact: I never knew how much a BOV affects the return to idle after revving. Every time I revved the car, it would dip super low when returning to idle, causing it to buck and stutter, and sometimes stall. Turns out this is because my BOV was set too "soft", and was actually cracking open slightly as the revs dropped down to idle, blowing off air and causing the dip in idle speed. A few turns to the "harder" setting, and it's spot-on perfect!

              Comment


                #8
                ^To all above it's pointless to tune the IAC controls until you can get it idling steadily with the ICV disabled.

                Originally posted by CorvallisBMW View Post
                On additional weird fact: I never knew how much a BOV affects the return to idle after revving. Every time I revved the car, it would dip super low when returning to idle, causing it to buck and stutter, and sometimes stall. Turns out this is because my BOV was set too "soft", and was actually cracking open slightly as the revs dropped down to idle, blowing off air and causing the dip in idle speed. A few turns to the "harder" setting, and it's spot-on perfect!

                http://www.turbosmartusa.com/technic...-bov-or-bpv-1/
                It has no effect at all on idle with a MAP based system unless your BOV piston has a leak.
                Last edited by varg; 04-06-2017, 09:29 AM.

                IG @turbovarg
                '91 318is, M20 turbo
                [CoTM: 4-18]
                '94 525iT slicktop, M50B30 + S362SX-E, 600WHP DD or bust
                - updated 1-26

                Comment


                  #9
                  Originally posted by varg View Post
                  ^To all above it's pointless to tune the IAC controls until you can get it idling steadily with the ICV disabled.
                  I have heard of doing it that way too, which I suppose might be easier in some ways, since you're tuning only 1 variable at a time.

                  Originally posted by varg View Post
                  It has no effect at all on idle with a MAP based system unless your BOV piston has a leak.
                  If the spring is set too soft, the piston lifts when it shouldn't, causing a "leak".

                  Comment


                    #10
                    Originally posted by CorvallisBMW View Post
                    If the spring is set too soft, the piston lifts when it shouldn't, causing a "leak".
                    This has no effect on idle with a MAP based system unless the BOV is broken and the vacuum side is leaking.

                    The only time this would have an effect is if your system has an airflow sensor before the blowoff valve.

                    IG @turbovarg
                    '91 318is, M20 turbo
                    [CoTM: 4-18]
                    '94 525iT slicktop, M50B30 + S362SX-E, 600WHP DD or bust
                    - updated 1-26

                    Comment


                      #11
                      Originally posted by varg View Post
                      This has no effect on idle with a MAP based system unless the BOV is broken and the vacuum side is leaking.

                      The only time this would have an effect is if your system has an airflow sensor before the blowoff valve.
                      Well IDK, but after following the instruction per the BOV manufacturer (Turbosmart) my dipping idle seems to be solved completely.

                      Comment

                      Working...
                      X